Porcine pancreatic amylase. We will try to delineate the substrate binding site by testing the role of OH groups in the substrate as H-bond donors or acceptors, by using amylose in which a very few of the OH groups have been replaced by H, H3, F or other small atoms or groups. Bacillus macerans 'amylase' (cyclotransglycosylase). We will attempt to separate transferase from hydrolytic activity. Possibly both activities are due to the same enzyme. In any case, with the pure transferase we will try to elucidate the mechanism, specifically to search for a covalent (glycoside ester?) intermediate and to determine to what extent the enzyme can act by an "endo" (as opposed to the conventional "exo") mechanism. Starch synthase. By using maize genetic variants, we wish to delineate amylose synthesis. Specifically, we suspect that this may involve unknown intermediates. Also we will try to see whether amylose is synthesized by an insertion mechanism at the reducing end (as opposed to 'conventional' synthesis at the non-reducing end).